Our major accomplishments this year are in the following areas: 1) Analyses of inner ears in Pax2 knockout mice Mutations in the PAX2 gene cause renal coloboma syndrome in humans that is associated with kidney abnormalities, optic nerve colobomas and mild sensorineural deafness. Pax2 knockout mice serve as a good animal model for the study of this syndrome since these mice display developmental defects in the same organs as human patients. We show that the overall development, outgrowth, and differentiation of the cochlear duct are affected in the Pax2 knockout mice. Even though the organ of Corti is poorly developed, sensory hair cells do form in the mouse mutants. These results raise the possibility that the etiology of hearing loss in renal coloboma syndrome may not be due to deficits within sensory hair cells but due to disruption of other cochlear functions. 2) Roles of FGF and BMP2 in the development of the apparatus that detects head movements The ability to detect angular head movements in vertebrates lies in the three semicircular canals and their sensory tissues, the cristae. The molecular mechanisms involved in the formation of this apparatus are not known. Using gain- and loss-of-function experiments in chicken, we show that FGFs in the presumptive cristae are important for crista formation. In addition, FGF emanating from the sensory regions promote canal formation by up-regulating Bmp2 expression in the canal genesis zone. Our results provide the first molecular evidence whereby sensory organs direct the development of their associated, non-sensory components, the semi-circular canals.